A percentile is a measure that indicates where a value ranks within its group.
Timeline uses the term 'percentile' to describe where a given outcome sits within the range of historical or illustrated scenarios.
To explore how a retirement income strategy might fare under a wide range of market conditions, Timeline runs multiple historical and/or Monte Carlo scenarios.
This means that there are hundreds and sometimes, thousands of possible outcomes. For instance, suppose we run every 30-year scenarios from Jan., 1900 to Dec., 2018, using monthly historical investment return and inflation. We'd have 1068 scenarios.
This means, if we want to look at legacy amount at the end of the 30-years, we'd have 1068 different amounts for all the 30-year rolling periods. (This is also true for income amounts by the way)
To make sense of the data, we rank all the amounts, into percentiles.
- 0th percentile is the lowest legacy amount i.e the worst-case scenario at the end of the 30-year period.
- 100th percentile amount is the highest i.e the best case scenario
- 50th percentile is the median amount. Half of all the other scenarios are better than the 50th percentile and half are worst.
- 20th percentile is scenario that ranks in the bottom 20 percent. This means that 80 out of every 100 scenarios are better than this one.
The default setting in Timeline is 10th percentile scenario. This outcome ranks in the bottom 10 percent, meaning that 90 out of every 100 scenarios are better than this. Or put another way, only 9 out of every 100 scenarios are worse than this.
Why 10th percentile?
This comes down to how we prefer to set client expectation. It's a very conservative estimate of the expected balance or income amount.
We know that people tend to anchor on amounts. So we default the software to an amount that is significantly lower than average.
Of course, users can chose their own preferred percentile ranking. If you want to present the worse case historical scenario to clients, then select the 0th percentile. If you want to present the average, select the 50th percentile. Or any other percentile that you prefer.
Real life example? Sure.
Suppose there are 20 people at your firm and you are fourth tallest person in that group, you are therefore at the 80th percentile!
This means that you are taller than 79% of people at your firm. Or put another way, only 2 out of every 10 people at your firm are taller than you.
Suppose Linda is the tallest person at the firm, she'll be in the 100th percentile. She is taller than everyone.
And your friend Jim is the shortest person at the firm, so he is technically in the 0th percentile. But you won't actually tell him that, would you?